Announcement

Collapse
No announcement yet.

GNU Binutils Prepares For Intel FRED/LKGS

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • GNU Binutils Prepares For Intel FRED/LKGS

    Phoronix: GNU Binutils Prepares For Intel FRED/LKGS

    The GNU Assembler "Gas" as part of the GNU Binutils collection has landed support for Intel FRED and LKGS instructions...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Is it just me or does this feel like a future vulnerability in the making?

    Comment


    • #3
      Originally posted by AusMatt View Post
      Is it just me or does this feel like a future vulnerability in the making?
      It is just you and FRED is a major improvement in design that helps security. Unless you think the crazy interrupt table designs from the 1980s are perfect and can't be improved.
      At some point, the house of cards begins to topple over. It’s no secret that the x86 processor architecture is almost aged enough to collect a pension. It is, not to put too fine a point on it, a c…

      Comment


      • #4
        Yeah, FRED sounds great. When I read the spec, it seemed to me that there's finally a foolproof way to tell the difference between an exception and an interrupt. Of course, anyone writing a 64-bit OS that can take advantage of FRED won't have mapped any HW interrupts in the first 0x20 vectors anyway (unlike IBM decided to do in the 80s), but still, better late than never!

        Comment


        • #5
          I'm waiting for the WILMA instruction.

          Comment


          • #6
            Originally posted by PluMGMK View Post
            Yeah, FRED sounds great. When I read the spec, it seemed to me that there's finally a foolproof way to tell the difference between an exception and an interrupt. Of course, anyone writing a 64-bit OS that can take advantage of FRED won't have mapped any HW interrupts in the first 0x20 vectors anyway (unlike IBM decided to do in the 80s), but still, better late than never!
            Do you mean telling the difference between explicitly calling interrupt on one of the interrupt codes for an exception and the hardware calling it? I was not aware that was an issue in practice.

            Also, here is a link to the specification for those who were curious:

            Comment


            • #7
              Well, that and telling the difference between IRQ5 and #GP, for example. Again, I know nobody actually maps IRQs to interrupts 8-10h anymore, and even in the old days there were ways to cope. So yeah, not (anymore) an issue in practice, but still, nice to see all the same!

              Comment

              Working...
              X